Hacker News Reader

I'd like to think there are some things today which we have got completely wrong. That we're not just refining correct theories.

I think today's explosion of LLMs etc will lead (eventually) to actual understanding. Data precedes theory.

You're thinking of nuclear waste? Not representative of today's time capsule, any more than space shuttle software development is representative of software development in general.

View on HN

"I think today's explosion of LLMs etc will lead (eventually) to actual understanding."

Or the opposite. Most people become too lazy to think and let the AI do it for them... and so the AIs can never overcome their baked in biases, so no real progress will happen.

And about what we have got wrong, I think the problem is mostly how we (the average human) lives.

Every story about dark energy should just be labeled “our astronomy has some large but unknown flaw.”

Reply on news.ycombinator.com